sign in sign up
<<
Home , Arctic wolf, Baffin Island, Ellesmere Island, Lowther Island, Parry Channel, Queen Elizabeth Islands

VOL. 48, NO.4 (DECEMBER 1995) P. 313–323

Wolf-Sightings on the Canadian Arctic Islands FRANK L. MILLER1 and FRANCES D. REINTJES1

(Received 6 April 1994; accepted in revised form 13 March 1995)

ABSTRACT. A -sighting questionnaire was sent to 201 arctic field researchers from many disciplines to solicit information on observations of ( lupus spp.) made by field parties on Canadian Arctic Islands. Useable responses were obtained for 24 of the 25 years between 1967 and 1991. Respondents reported 373 observations, involving 1203 wolf-sightings. Of these, 688 wolves in 234 observations were judged to be different individuals; the remaining 515 wolf-sightings in 139 observations were believed to be repeated observations of 167 of those 688 wolves. The reported wolf-sightings were obtained from 1953 field-weeks spent on 18 of 36 Arctic Islands reported on: no wolves were seen on the other 18 islands during an additional 186 field-weeks. Airborne observers made 24% of all wolf-sightings, 266 wolves in 48 packs and 28 single wolves. Respondents reported seeing 572 different wolves in 118 separate packs and 116 single wolves. Pack sizes averaged 4.8 ± 0.28 SE and ranged from 2 to 15 wolves. Sixty-three wolf pups were seen in 16 packs, with a mean of 3.9 ± 2.24 SD and a range of 1–10 pups per pack. Most (81%) of the different wolves were seen on the . Respondents annually averaged 10.9 observations of wolves ·100 field-weeks-1 and saw on average 32.2 wolves·100 field-weeks-1· yr -1 between 1967 and 1991. Average rates of wolf observations·100 field-weeks-1 (28.5, 13.6 vs. 5.7; p < 0.005) and mean numbers of different wolves seen·100 field-weeks-1 (92.3, 37.5 vs. 15.4; p < 0.005) were markedly greater during 1967–75 and 1989–91 than in 1976–88. Relative differences in the reported rates of wolf observations on the Queen Elizabeth Islands in 1967– 75, 1976–88, and 1989– 91 follow the relative abundance of the wolf’s major prey, (Rangifer tarandus pearyi) and muskoxen (Ovibos moschatus), on those islands during those periods. Key words: arctic-island wolves, Canis lupus spp., Canadian Arctic Islands, wolf-sighting questionnaire, reported wolf observations

RÉSUMÉ. Un questionnaire portant sur l’observation de loups a été envoyé à 201 chercheurs de divers domaines travaillant dans l’Arctique, dans le but de solliciter de l’information sur la présence de loups (Canis lupus sp. plur.) notée par des groupes sur le terrain dans l’archipel Arctique canadien. Des réponses utilisables ont été obtenues pour 24 des 25 années allant de 1967 à 1991. Les répondants ont rapporté 373 cas d’observations au cours desquels ils ont vu 1203 loups. De ces derniers, 688 loups ont été jugés être des individus distincts, comptés lors de 234 cas d’observations; parmi les autres 515 loups aperçus au cours des 139 autres cas d’observations se retrouvaient sans doute 167 des 688 loups mentionnés ci-dessus. Les observations de loups rapportées ont été faites durant 1953 semaines de travail sur le terrain passées sur 18 des 36 îles de l’archipel Arctique sur lesquelles portait la recherche; aucun loup n’a été vu sur les autres 18 îles au cours de 186 semaines de travail supplémentaires sur le terrain. Vingt- quatre p. cent des cas d’observations de loups ont été faits par des observateurs aéroportés, qui ont noté 226 loups en 48 bandes et 28 loups solitaires. Les répondants ont dit avoir vu 572 loups différents répartis en 118 bandes distinctes et 116 loups solitaires. La taille des bandes était en moyenne de 4,8 ± 0,28 ES et allait de 2 à 15 loups. Soixante-trois louveteaux ont été aperçus dans 16 bandes ayant de 1 à 10 louveteaux par bande et une moyenne de 3,9 ± 2,24 EMQ louveteaux par bande. La plupart (81 p. cent) des loups différents ont été aperçus sur les îles de la Reine-Élisabeth. Les répondants ont vu des loups en moyenne 10,9 fois par an·100 semaines sur le terrain-1 et ont compté en moyenne 32,2 loups·100 semaines sur le terrain-1·an-1 entre 1967 et 1991. La fréquence moyenne d’observations de loups·100 semaines sur le terrain-1 (28,5, 13,6 c. 5,7; p < 0,005) et le nombre moyen de loups différents aperçus·100 semaines sur le terrain-1 (92,3, 37,5 c. 15,4; p < 0,005) étaient nettement plus élevés au cours des années 1967 à 1975 et 1989 à 1991 qu’au cours des années 1976 à 1988. Les différences relatives entre les fréquences d’observations de loups rapportées sur les îles de la Reine-Élisabeth en 1967-1975, 1976-1988 et 1989-1991 s’accordent avec l’abondance relative des proies principales du loup, le caribou de Peary (Rangifer tarandus pearyi) et le boeuf musqué (Ovibos moschatus), sur ces îles au cours de ces périodes. Mots clés: loups de l’archipel Arctique, Canis lupus sp. plur., archipel Arctique canadien, questionnaire portant sur les cas d’observations de loups, observations de loups rapportées

Traduit pour la revue Arctic par Nésida Loyer.

1 Canadian Wildlife Service, Northern Conservation Division, Environmental Conservation Branch, Prairie & Northern Region, Environment , Room 200, 4999 – 98 Avenue, , Alberta T6B 2X3, Canada © The Arctic Institute of 314 • F.L. MILLER and F.D. REINTJES

INTRODUCTION archipelago (Babb and Bliss, 1974; Edlund and Alt, 1989; Bliss, 1990; Edlund, 1990). For the purpose of reporting our Wolves (Canis lupus spp.) on the Canadian Arctic Archi- findings, we have divided the CAA into three regions and ten pelago (CAA) are highly visible on the open tundra to ground major areas generally reflective of climate and patterns of observers and to relatively low-level airborne observers, vegetation cover (cf. Miller, 1993) (Figs. 1 to 3). The “Queen when the wolves are moving. Visibility is greatest when the Elizabeth Islands region” (QEI) is divided into five major wolves are on a snow-free background, which generally areas: (1) eastern islands, (2) southwestern islands, (3) south- corresponds to the period of maximal human activity on the central islands, (4) north-central islands, and (5) northwest- CAA, late June to August. Adult wolves on the Arctic Islands ern islands. The “Southern Tier of Arctic Islands region” (STI) usually are white or generally white or whitish in appearance. includes three major areas: (1) western islands, (2) eastern is- They are curious and most are relatively bold compared to lands, and (3) south-central islands. The “ region” other subspecies of C. lupus, and often behave like shy dogs (BIR) is divided into (1) northern and (2) southern areas. in the presence of humans (e.g., Grace, 1976; Miller, 1978; Munthe and Hutchinson, 1978). Arctic-island wolves tend to investigate everything that is new to them, including field MATERIALS AND METHODS camps or field party members that they encounter in their environment. Thus, if wolves are present in the area of a field Three subspecies of the North American gray wolf have camp, they are likely to be seen. been identified on the CAA: (1) C. l. arctos, Pocock (1935), Essentially all of the ca. 11 500 year-round residents on the originally referred to as the “American tundra wolf” by CAA ( Data Book 1990/91, 1990) live Pocock (1935) but subsequently called the Canadian polar in widely scattered, small, coastal settlements. Even wolf by Anderson (1946); (2) C. l. manningi, Anderson though Inuit hunters and trappers on occasion travel over (1943), the Baffin Island tundra wolf; and (3) C. l. bernardi, large areas of the Arctic Islands (e.g., Riewe, 1992), vast Anderson (1943), the tundra wolf. C. l. arctos tracts of land remain isolated and remote, except for infre- occurs throughout the QEI (those islands north of the Parry quent visits or overflights by scientific or exploration field Channel or north of 74˚N latitude, excluding the BIR) and parties. Therefore, most arctic-island wolves apparently lack most likely throughout the southern tier of islands (those experience with humans, especially those wolves on the islands south of the or south of ca. 74˚N Queen Elizabeth Islands. latitude), except Baffin and its satellite islands (Miller, 1993). There is little information on the biology, ecology, popu- Pelage descriptions of the three arctic-island subspecies have lation dynamics or current status of wolves on the CAA. A been made by Pocock (1935), Anderson (1943), Manning recent evaluation has been made, however, of the probable (1943), Manning and Macpherson (1958, 1961) and Clark status of wolves on the Arctic Islands by examining the (1971). There is likely intergradation of C. l. arctos with “theoretical maximum carrying capacities” for wolves in mainland subspecies such as C. l. hudsonicus and C. l. various areas of the archipelago in relation to their available mackenzii on the STI and, at least sporadically, between C. l. ungulate prey base (Miller, 1993). The ungulate prey base is arctos on and C. l. orion from or composed solely of caribou (Rangifer tarandus spp.) and on Greenland (e.g., Manning and Macpherson, 1961; Dawes muskoxen (Ovibos moschatus). Wolves are large, attractive et al., 1986). C. l. manningi is known only from Baffin Island; carnivores that are of interest to people and usually of strong but most recently, R.M. Nowak, (Staff Specialist, Office of interest to arctic field researchers, regardless of their disci- Endangered Species, U.S. Fish and Wildlife Service, pers. pline. Therefore, in the absence of essentially any systematic comm. 1992) has suggested that C. l. manningi should prob- and quantitative data on past, recent, or current numbers of ably be lumped into a major North American group—C. l. wolves on the CAA, we produced a “wolf-sighting” question- nubilus— along with the mainland races of C. l. hudsonicus naire to solicit observations of wolves made by field parties and C. l griseoalbus. C. l. bernardi is known only from its on the CAA. Herein, we report our analyses of the data type locality in the Cape Kellett area of Banks Island. It obtained from the wolf-sighting questionnaire. appears to have gone extinct sometime between 1918 and 1952, or the specimens examined were merely immigrants from the mainland and did not represent a valid subspecies STUDY AREA (cf. Manning and Macpherson, 1958). In 1990, we mailed a wolf-sighting questionnaire to 201 The CAA forms the northern apex of North America and arctic field researchers soliciting the following information: encompasses a landmass of more than 1.3 million km2. It (1) year of observation, (2) month(s) of observation, (3) represents one-seventh of the land area of Canada and spans week(s) spent in the field, (4) island where sighting(s) was nearly 60˚ of longitude (66–126˚ W) and 22˚ of latitude (61– made, (5) wolf pack size (given by 1+ yr old wolves plus 83˚ N). The tundra vegetation of the Canadian Arctic Islands pups), (6) number of solitary individual wolves seen, (7) consists of , bryophytes, graminoids, herbs, cushion number of times the same wolf or wolves were seen, (8) plants, and prostrate shrubs with some low, erect shrubs occur- whether the observer(s) was airborne or on the ground, and ring on protected sites in southern and eastern parts of the (9) the name of the observer. We received 99 (49%) responses ARCTIC ISLAND WOLF-SIGHTINGS • 315

FIG. 1. Queen Elizabeth Islands region (QEI) divided into five major areas by apparent importance to arctic-island wolves. to the 201 questionnaires but data from only 78 of those a field party, with no actual measure of the number of people respondents were usable. The other 21 respondents simply involved, the proportion of time spent in camp vs. out of informed us that they had no data on the subject (for a wide camp, the type of work being carried out, or the proportion of variety of reasons), or they reported only for areas other than time spent airborne vs. working on the ground. We recognize the CAA. Subsequently, we received completed wolf-sight- that there are unmeasured (and possibly unknown) factors ing questionnaires from an additional eight field researchers that would or could have influenced the accuracy of this for the 1991 only. measure of observer effort based on our field-week unit. It is, It was not practical to expect field researchers to provide however, the most feasible unit that we could obtain to help the degree of detail most desirable for a highly accurate reflect the “commonness of wolf sightings over time” from measure of “search effort” or the “rate of wolf-sightings” people’s recollections and records of field activities on the (i.e., number of person-hours in the field under conditions Arctic Islands over the past four decades. that likely would have been associated with a high probability Respondents to our wolf-sighting questionnaire reported of making a wolf-sighting). Therefore, we opted for a general on 545 field periods (2206 field-weeks) on 36 Canadian measure of “observer effort,” which we termed a “field- Arctic Islands: 14 major islands (> 10 000 km2), 10 interme- week,” to approximate the effort involved in obtaining wolf- diate-sized islands (> 1000 to < 10 000 km2), and 12 small sightings (single wolves or wolf packs). We define a “field- islands (< 1000 km2). Their responses included information week” as one week of field activity by one or more persons in on field periods during 31 of the 41 years between 1950 and 316 • F.L. MILLER and F.D. REINTJES

FIG. 2. Southern Tier of Arctic Islands region (STI) divided into three major areas by apparent importance to arctic-island wolves.

1991, with no information given for 1952, 1954–60, 1964, to 1966 is both spatially and temporally fragmentary and, and 1969. The first sighting of a wolf was not reported, based on 1967–91 sighting rates and the relatively few weeks however, until 1967. We define a “wolf-sighting” as the spent on each island between 1950 and 1966, no more than sighting of one wolf on one occasion. For example, one wolf two wolf-sightings would have been expected from any of seen five times equals five wolf-sightings, five wolves seen those islands. Therefore, we have omitted those five responses once also equals five wolf-sightings, five wolves seen three for the years prior to 1967 from further consideration. times equals 15 wolf-sightings, and so on. We have treated repeat observations of wolves separately, Wolves were not seen by the five respondents (14 field as it is likely that with time the making of subsequent repeat periods) who collectively spent 67 field-weeks from June to wolf-sightings (multiple sightings of the same wolves) usu- September during seven years between 1950 and 1966 on ally will be markedly greater than making the first observa- seven different islands. These were Cornwallis Island (26 tion of those wolves. This condition pertains because field-weeks), 1950 (7), 1951 (8), 1953 (7), and 1961 (4); “resighting curves” generally decline with time for initial (15 field-weeks), 1963 (3), 1965 (4), and 1966 sightings of new individuals (or packs). For example, once (8); (12 field-weeks), 1965; Banks Island (6 arctic-island wolves have located a field camp, they are likely field-weeks), 1965; Bathurst Island (3 field-weeks), 1961; Bylot to return to it out of curiosity, as it represents a novel stimulus Island (3 field-weeks), 1963; and southern Baffin Island (2 in their environment. Such visits are likely to be frequent, field-weeks), 1965 (1), and 1966 (1). The information for 1950 particularly if the wolves experience no harmful events while ARCTIC ISLAND WOLF-SIGHTINGS • 317

FIG. 3. Baffin Island region (BIR) divided into two major areas by apparent importance to arctic-island wolves. visiting (that is, if the field party does not discourage the wolf wolves only. The following results are based on the informa- visits). Also, the field work being carried out may take tion obtained from 81 respondents, who provided records for members of the field party back to sites or areas where wolves 531 field periods, involving 2139 field-weeks on 36 Arctic were previously seen. Knowledge of the probable presence of Islands within the CAA for 24 of the 25 years (no data in 1969) wolves will often make field workers more alert to the from the first wolf-sighting in 1967 through summer 1991. possibility of seeing a wolf or wolves, which should increase the probability of doing so. Therefore, any resultant rate of wolf observations, wolf-sightings, or any measure of ob- RESULTS AND DISCUSSION server effort will most probably be markedly inflated by the relative contribution of repeat sightings of the same wolves to Respondents reported 373 observations, involving 1203 the contribution made by the first-time sightings of different wolf-sightings of which 688 wolves in 234 observations were 318 • F.L. MILLER and F.D. REINTJES judged to be different individuals and the remaining 515 condition, seemingly, suggests that single wolves are more wolf-sightings in 139 observations were repeated observa- difficult to detect from the air than are packs of two or more tions of 167 of those 688 wolves (Tables 1–3, Figs. 1–6). The wolves. Although it seems evident that the detection of reported wolf-sightings were obtained from 1953 field-weeks several wolves should be easier than detecting just one, it is spent on 18 Arctic Islands. In actuality, respondents made all also possible that once one wolf is seen, the observer is alerted of their wolf-sightings during 655 of those 1953 field-weeks to the possible presence of other wolves and is more likely to and they saw no wolves during the remaining 1298 field- detect others, if present, after the first wolf is seen. weeks that they spent on those 18 islands. Therefore, on islands where wolves were seen, respondents spent on aver- Wolf Packs age 8.3 field-weeks to make one observation of wolves (involving different wolves only) or 5.2 field-weeks for an Wolf packs were observed on 16 islands. Respondents observation that sometimes included repeat observations of reported seeing 572 different wolves in 118 separate packs some of the same wolves. Those observations averaged 35.2 (Table 1; Fig. 4: overall mean pack size 4.8 ± 0.28 SE, range different wolves ·100 field-weeks-1 or 61.6 wolf-sightings ·100 of pack sizes 2–15). Excluding the 30 pairs of wolves, the field-weeks-1 of field activities. No wolves were seen during resultant 88 packs averaged 5.8 ± 0.31 SE in size and ranged 186 field-weeks of collective field activities on the northern from 3 to 15 . The greatest number of packs was seen half of Baffin Island (55 field-weeks) and 18 other islands: on the QEI, 82% of all packs, with 80% of all different wolves Bylot, 57 field-weeks; Dundas, 30; Rowley, 16; King William, seen in packs, and 67% of all the different wolves that were 5; Griffith, 4; Baillie-Hamilton, 3; Ellef Ringnes, 3; Jenny reported. Within the QEI, 89% of the packs were equally Lind, 2; Prince Charles, 2; and one field-week each on distributed between the eastern and southwestern areas. Ob- Airforce, Bray, Brock, Cornwall, Crown Prince Frederick, servations from the STI contributed 15% of all the packs, with Foley, Meighen, Russell, and Truro. On the basis of the mean 18% of all the wolves seen in packs, and 15% of all the rate of sighting of one wolf per three field-weeks, 5–19 wolf- different wolves reported. Those packs were nearly equally sightings should have been made on Rowley, Dundas, north- distributed between the western (56%) and the eastern (44%) ern Baffin, and Bylot islands. Thus, the total lack of wolf- areas of the STI. Only three different packs (3%), with three sightings on those islands suggests that wolves were absent or wolves each, were seen in the BIR (on southern Baffin Island rare. The time spent on each island was too short to determine in three different years). relative density of wolves on any of the other 15 islands where There was no significant difference in the mean size of no wolves were seen. packs within the eastern (mean 4.2 ± 0.34 SE), southwestern Over four-fifths of all the observations (n = 234) of (5.4 ± 0.54), and south-central (4.3 ± 0.87) areas of the QEI; different wolves seen in packs or as single wolves were made by respondents on the QEI (Table 1). The largest number of wolves seen was on southwestern QEI. No wolves were reported for the northwestern QEI, the south-central islands of the STI, or the northern area of the BIR. Airborne observers made 20% (76) of all observations, and only 3% of those observations involved repeat sightings (three repeats of the same pack of 10 wolves during a single field period). Those 76 observations provided 24% (294) of all of the wolf-sightings. They included 266 sightings of wolves in 48 packs (22% of all wolf-sightings; 33% of all 146 packs, mean pack size 5.5 ± 0.49 SE, range of pack sizes 2–15) and 28 sightings of single wolves. The 73 first-time observations made from the air (31% of all observations of different wolves) contributed 38% (264) of all of the different wolves seen, including 236 wolves in 45 separate packs (41% of all different wolves seen in packs, 38% of all 118 different packs seen, mean pack size 5.2 ± 0.49 SE, range of pack sizes 2–15); and 24% (28) of all single wolves seen. Proportional representation of observations of wolves in packs vs. observations of single wolves varied significantly between airborne and ground observers (χ2 = 5.3, df = 1; p < 0.05). Compared to the number of separate packs (73) and single wolves (88) reported by ground observers, the number of observations made by airborne observers of wolves in FIG. 4. Distribution of wolves seen in packs during initial observations, as separate packs (45) was overrepresented and the number of indicated by the frequency of packs of different size, Canadian Arctic observations of single wolves (28) underrepresented. This Archipelago, 1967– 91. ARCTIC ISLAND WOLF-SIGHTINGS • 319

TABLE 1. Distribution of “observer effort” (n = 1953 field-weeks) and wolf-sightings of 688 different wolves by region, area, and island, Canadian , 1967–91.

Effort Number of Number of number of field-weeks different Number of Number Time years when Region/area when wolves were wolves Packs seen singletons of pups span wolves Island Seen Not seen seen n Mean ± SE Range seen seen involved were seen

Queen Elizabeth Islands region Eastern area: Ellesmere 196 198 197 34 4.4 ± 0.41 02–11 47 15 1967–91 19 Devon 27 237 8 2 3.5 ± 2.12 02–05 1 0 1967–88 3 Axel Heiberg 30 67 30 7 3.3 ± 0.61 02–06 7 0 1981–90 6 Southwestern area: Melville 84 24 181 27 5.8 ± 0.71 02–15 24 14 1971–88 10 Prince Patrick 67 12 70 12 5.2 ± 0.82 02–10 8 9 1973–89 6 Eglinton 12 0 13 3 4.0 ± 1.00 02–06 1 1 1973–74 2 Byam Martin 16 0 3 1 2.0 ± 0.00 02–02 1 0 1973–74 2 South-central area: Bathurst 31 38 30 6 4.0 ± 1.37 02–10 6 0 1973–90 7 Cornwallis 6 233 8 1 6.0 ± 0.00 06–06 2 0 1970–89 2 Cameron 3 1 13 3 4.3 ± 1.20 02–06 0 3 1988–91 3 Lowther1 3 1 1 0 – – 1 0 1981 1 North-central area: Lougheed 3 0 1 0 – – 1 0 1979 1 King Christian 2 5 3 1 3.0 ± 0.00 03–03 0 1 1982 1

Southern Tier of Arctic Islands region Western area: Victoria 5 96 14 3 4.7 ± 1.20 03–07 0 2 1987–90 2 Banks 50 139 41 7 5.4 ± 1.29 02–10 3 10 1974–90 8 Eastern area: Prince of Wales 57 24 52 7 7.1 ± 1.28 02–11 2 8 1976–89 6 Somerset 16 105 4 1 2.0 ± 0.00 02–02 2 0 1988–91 3

Baffin Island region Southern area: South Baffin 47 118 19 3 3.0 ± 0.00 03–03 10 0 1967–90 6

1 was placed in the south-central area of the Queen Elizabeth Islands region although it lies to the south in the waters of .

and the western (5.2 ± 0.94), and eastern (6.5 ± 1.28) areas of September of the year. Most observers, however, would not the STI. Sample sizes for packs in the north-central area (n = 1) necessarily recognize the previous year’s pups, when seen of the QEI and the southern area (n = 3) of the BIR were too during the months from January through May. The propor- small to evaluate for comparison from area to area. The number tion of packs with pups present is still only 17%, when the 26 of wolves seen in packs on an island was strongly correlated packs (167 wolves, mean pack size 6.4 ± 3.80 SD, range of with the number of packs seen (Pearson Correlation, r = 0.982; pack sizes 2–15) seen between January and May are deleted p = 0.001) but not with mean pack size (Pearson Correlation, from this consideration. The proportion of packs with pups r = 0.453; p = 0.078), suggesting that mean pack size was present can be increased to a maximum of 24% (16 of 66 similar among all islands where wolf packs were seen. packs) by deleting the 30 pairs of wolves included in the pack data and the 22 packs of three or more wolves each seen Wolf Pups sometime during January through May of the year. These data cannot be used directly to measure pup produc- Respondents saw 63 pups in 16 packs (Table 1, Fig. 5). Mean tivity among years, as arctic-island wolf pups are not regular pack size was 7.9 ± 3.14 SD; range, 3–15, with a mean of travellers with the packs until essentially the beginning of 3.9 ± 1.77 SD and a range of 1–7 for 1+ yr olds and a mean winter. Arctic-island wolf pups continue to use rendezvous of 3.9 ± 2.24 SD and a range of 1–10 for pups. The most fre- sites within their summer homesite (Clark, 1971; Packard et quently observed numbers of pups in a pack were four and five al., 1992) or recent kill sites as temporary rendezvous sites, (four observations each). Significantly (χ2 = 10.56, df = 1; when they begin to travel with the pack after leaving the p < 0.005) fewer 1+ yr old wolves were present in packs with summer homesite (Gray, 1993) during August or even through 5–10 pups than in packs containing only 1– 4 pups (Fig. 5). September of the year. Also, females with pups are wary and Wolf pups were seen in only 14% of the 118 packs may not approach close enough to readily be seen by people reported. All of the pups were seen between June and working at other activities. We believe, however, that the 320 • F.L. MILLER and F.D. REINTJES

TABLE 2. Distribution of 515 wolf-sightings from repeat observations of 167 of the 688 different wolves by region, area, and island, Canadian Arctic Archipelago, 1967–91.

Packs Number Total Region/area Number of Number of repeat Number of repetitions repetitive Island singletons n of wolves observations Involving packs Involving singletons wolf-sightings

Queen Elizabeth Islands region Eastern area: Ellesmere 17 11 56 60 24 36 180 Axel Heiberg 3 5 14 16 11 5 33 Southwestern area: Melville 2 5 24 19 17 2 96 Prince Patrick 2 2 10 19 11 8 63 South-central area: Bathurst 2 0 0 2 0 2 2 North-central area: King Christian 0 1 3 1 1 0 3

Southern Tier of Arctic Islands region Western area: Victoria 0 1 4 1 1 0 4 Banks 2 1 10 11 9 2 92 Eastern area: Prince of Wales 1 2 16 9 4 5 41

Baffin Island region Southern area: South Baffin 1 0 0 1 0 1 1 apparent lack of pups in some years is likely real. most of the single wolves seen were simply temporarily Most of the pups seen (43 in 11 observations, 68% of all absent “pack wolves” on solitary foraging trips. pups seen) were on the QEI and the remaining pups (20 in five observations, 32% of all pups seen) were on the STI (Table 1). Repeat Observations No wolf pups were reported for the BIR. The greatest number of pups in a single pack (n = 10) occurred on Melville Island. Respondents made 139 repeat observations of wolves (Table 2: mean number of repeat observations 2.5 ± 0.32 SE, Single Wolves range 1–9 times each). Those repeat observations involved 28 different packs (137 different wolves, mean pack size Single wolves were reported for 15 islands, and respond- 4.9 ± 0.50 SE, range of pack sizes 2–11) during 78 repeat ents saw 116 different single wolves (Table 1: 17% of all 688 events (mean number of repeat observations per pack wolves seen). Most (85%) of those single wolves were seen on the QEI, followed by 9% seen in the BIR, and 6% on the STI. Single wolves were reported for only four of the five major areas within the QEI. The largest number of single wolves on the QEI was seen on eastern islands (47% of all singletons), followed by those on southwestern islands (29%), south-central islands (8%), and a north-central island (ca. 1%). All 10 of the single wolves seen in the BIR were in the southern area. Most of the single wolves within the STI were seen on eastern islands (ca. 3%), followed by the three remaining singletons (ca. 2%) on a western island, and no singletons were seen on any of the south-central islands. The number of single wolves seen on an island basis was directly related to the number of wolf packs seen (Pearson Correlation, r = 0.942; p = 0.001) and the total number of wolves present in packs (Pearson Correlation, r = 0.876; p = 0.001). There was no relationship, however, between mean pack sizes and the number of single wolves seen on an island basis (Pearson Correlation, r = 0.145; p = 0.592). Thus, FIG. 5. Total number of wolf pups and 1+ yr old wolves in packs with pups the existing evidence supports the possibility that many or present as a function of number of pups in the pack, Canadian Arctic Archipelago, 1967–91. ARCTIC ISLAND WOLF-SIGHTINGS • 321

TABLE 3. Chronological distribution of observational effort, wolf-sightings, and observational success for 24 years when field parties reported seeing wolves, Canadian Arctic Archipelago, 1967–91.

Wolf Observations Repetitive wolf-sightings Year1 Field-weeks Number of Total Number Number of Number of Number Total (n) observations wolves of Packs Singletons observations of wolves sightings

1967 42 8 27 622612 1968 21 6 10 24000 1970 21 3412000 1971 32 8 13 17000 1972 50 26 80 11 15 6 13 23 1973 56 19 77 12 7000 1974 75 21 90 14 7 10 11 67 1975 40 5 10 14212 1976 60 4 10 2 2 11 7 19 1977 97 5 19 3 2 16 8 55 1978 55 3721000 1979 58 3 13 12111 1980 107 5 23 4 1 3 10 30 1981 111 3303111 1982 98 4 12 22133 1983 81 2 10 20000 1984 103 3 11 21111 1985 113 9 18 455812 1986 124 8 10 26355 1987 161 12 43 9 3 6 14 20 1988 186 16 30 7 9 22 21 115 1989 180 26 84 16 10 5 9 10 1990 215 18 57 9 9 17 29 98 1991 53 17 27 5 12 27 19 41

1 No information was reported for 1969.

2.8 ± 0.48 SE, range 1–9 times each) and 30 single wolves wolves reported was highly variable from year to year (Table during 61 repeat events (mean number of repeat observations 3): total number averaged 28.7 ± 5.65 SE (range 3–90); per individual 2.0 ± 0.33 SE, range 1–5 times each). All of number of wolf packs averaged 4.9 ± 0.93 SE (range 0–16); the repeat observations were obtained by ground observers, and single wolves averaged 4.8 ± 0.80 SE (range 0–15). No with the exception of three repeat observations of the same positive relationship could be found between observer effort pack of wolves during one field period while members of the (the number of field-weeks per year) and the number of field party were airborne. wolves seen annually (Pearson Correlation, r = 0.295; Most repeat wolf observations and repetitive wolf- p = 0.161) or the mean number of wolves per pack (Pearson sightings were from field activities on the QEI (Table 2). Correlation, r = 0.081; p = 0.080). The remaining repeat wolf observations and repetitive The temporal distribution of the wolf observations (Table wolf-sightings, with the exception of one of each, came 3, Fig. 6) suggests, however, that wolves were both more from the STI. The number of wolf-sightings per unit of common and seen at greater rates during the periods 1967–75, effort (field-weeks) increased threefold once the initial and 1989–91 than during 1976–88. The annual mean sighting was made (515 repeats vs. 167 initial), suggesting number of wolf observations·100 field-weeks-1 exceeded the that wolves tended to remain in the vicinity of observers overall mean (10.9, 1967–91), most frequently in 1967–75, following initial contact, at least during the 178 field- followed by 1989–91, and remained continually below the weeks of observations when repeat sightings were made. overall mean throughout 1976–88. The number of wolves The amount of effort required to make an original sighting seen·100 field-weeks-1·yr-1 (32.2, 1967–91) followed essen- was much greater than that required for subsequent tially the same pattern. resightings. This relative preponderance of repeat sightings The numbers of wolf observations and wolves seen were over the initial observation supports the assumption that significantly greater in 1967–75 and 1989–91 than in 1976–88 the combination of all sightings (repeat + original) would (χ2 = 498.0, df = 2, p < 0.005; and χ2 = 131.8, df = 2, p < 0.005, have led to an unjustified inflation of the apparent yield per respectively). Wolf observations averaged 28.5 observa- unit of observer effort. tions· 100 field-weeks-1 in 1967–75, 13.6 in 1989–91, and 5.7 in 1976–88; while mean numbers of wolves seen averaged Temporal Considerations 92.3 wolves·100 field-weeks-1 in 1967–75, 37.5 in 1989–91, and 15.4 in 1976–88. Respondents averaged 10.9 observations of wolves annu- Packs of wolves were reported for 23 of the 24 years when ally and saw a mean of 32.2 wolves ·100 field-weeks-1· yr-1 wolves were seen between 1967 and 1991 (Table 3). The between 1967 and 1991 (Table 3, Fig. 6). The number of greatest number of wolf packs was seen in 1989 (14% of all 322 • F.L. MILLER and F.D. REINTJES

however, that on average, respondents had about a one-in- three chance of making a wolf observation during any one week of field activity on the Arctic Islands where wolves were seen and reported between 1967 and 1991.

Relation to Ungulate Prey

The resultant curves (Fig. 6) are interesting in the light of known changes in numbers of Peary caribou (Rangifer tarandus pearyi) and muskoxen (Ovibos moschatus) on the QEI from 1961 to 1994 (e.g., Tener, 1963; Miller et al., 1977; Miller, 1990, 1993, unpubl. data 1984–94). Both Peary caribou and muskoxen declined drastically in the winter of 1974–75. Peary caribou numbers then continued downward during the late 1970s and into the early 1980s. A slight increase in the number of Peary caribou on the south-central QEI was first detected in the mid 1980s, with a markedly noticeable increase occurring there from the late 1980s into FIG. 6. Annual rates of reported wolf observations and wolf-sightings, Canadian Arctic Archipelago, 1967– 91. the early 1990s. Muskoxen were quicker to respond and began their recovery with a slow increase in the early 1980s, packs, totalling 74 wolves), followed by 1974 (12% of all if not sooner. They then appeared to increase range-wide by packs, totalling 83 wolves). the mid to late 1980s and into the 1990s. Thus, the reported Respondents saw 63 wolf pups in 16 packs during 9 of the rates of wolf observations and wolf-sightings followed the 23 years when wolf packs were seen (no packs were observed relative abundance of their major prey on the QEI—the Peary in 1981). The largest number of pups was observed in 1973 caribou and the . It is impossible, however, to propose (19 pups in five observations, 30% of all pups seen); followed with confidence that the distributions presented in Figure 6 by 1989, when 17 pups were observed in four sightings (27% necessarily document a cyclic predator-prey relationship of all pups seen). Wolf pups were never seen in more than two among wolves, Peary caribou, and muskoxen on the QEI consecutive years (1973–74, 1977, 1979, 1982–83, 1987, from 1967 to 1991, because too many unknowns are associ- and 1989–90), with as many as five consecutive years ated with the collection of these reported data. (1967–72) passing without any sightings of pups being Comparisons of the distributions of wolves seen and reported (the degree that the data for observation of pups, or reported by questionnaire (Table 1) are only in partial agree- the lack of such data, is influenced by pups’ using rendezvous ment with the recently estimated availability of potential sites while the packs are travelling remains unknown). ungulate prey among the three regions and ten major areas of Respondents reported seeing single wolves in 23 of the 24 the CAA (cf. Miller, 1993). When the QEI region is consid- years between 1967 and 1991 (Table 3). The largest number ered by itself, the reported distribution of the relative number of different single wolves seen annually was in 1972 (13% of of wolves seen, the rate of wolf observations, and the rate of all singletons), followed by 1989 (10% of all singletons), and wolf-sightings are all in close agreement with the estimated in 1988 (9% of all singletons). The remaining 68% (79 single- sizes of the potential ungulate prey base within each of the tons) of the 116 different single wolves reported were con- five major areas (cf. Miller, 1993). Of the different wolves tributed over 20 years at annual rates of ca. 1% to 8% (1– 9 seen in the QEI, 90% were reported on eastern and southwest- singletons, yearly). ern islands (Table 1) and 83% of the potential ungulate prey We cannot evaluate the observed annual variation in the base on the QEI is estimated for those islands (Miller, 1993). reported wolf observations and wolf-sightings between 1967 When we make further intra- and inter-regional comparisons, and 1991 (Table 3, Fig. 6) with a high degree of confidence. however, the plausibility of the representative accuracy of the That is, we have no way of truly determining how the reported wolf observations must be questioned in relation to observers involved (e.g., their training, experience and inter- the relative availability of ungulate prey. est), location of field activities (island, area of an island, weather, time of year), and the type of field work being (1) Why is the reported rate of wolf observations in the QEI carried out (wide-ranging or in close to camp, and particularly about threefold greater than in the BIR and STI, when the low-level airborne vs. ground activities) influenced the prob- estimated ungulate prey base is tenfold and fivefold as ability of making, recording (or remembering), and reporting great, respectively, in the BIR and STI as in the QEI (Miller, wolf-sightings on our questionnaire. Also, marked changes in 1993: QEI, 20 500; BIR, 210 000; and STI, 111 300)? the size or availability of the ungulate prey base over time within the different regions or major areas of the CAA where (2) Why is the reported rate of wolf-sightings over threefold field activities were carried out would have influenced wolf greater for the STI than for the BIR, when the estimated numbers and, thus, the probability of seeing a wolf. It appears, ungulate prey base is essentially twice as great in the ARCTIC ISLAND WOLF-SIGHTINGS • 323

BIR as in the STI (Miller, 1993: BIR, 210 000; and STI, CLARK, K.R.F. 1971. Food habits and behaviour of the tundra 111 300)? wolf on central Baffin Island. Ph.D. thesis, University of Toronto, Toronto, Ontario. 223 p. (3) Why is the reported rate of wolf-sightings for the eastern DAWES, P.R., ELANDER, M., and ERICSON, M. 1986. The wolf islands of the STI 42% greater than the rate for the (Canis lupus) in Greenland: A historical review and present western islands, when the estimated ungulate prey base status. Arctic 39:119–132. is over eightfold as great on the western islands (Miller, EDLUND, S.A. 1990. Bioclimatic zones in the Canadian Arctic 1993: western STI, 99 000 vs. eastern STI, 12 000)? Archipelago. In: Harington C.R., ed. Canada’s missing dimension—science and history in the Canadian Arctic Islands, (4) Why were no wolves seen in the northern area of the BIR, Volume 1. Ottawa: Canadian Museum of Nature. 421–441. when there are fourfold as many ungulates (barren- EDLUND, S.A., and ALT, B.T. 1989. Regional congruence of ground caribou, R. t. groenlandicus) estimated for the vegetation and summer climate patterns in the Queen Elizabeth northern area of the BIR as for the entire QEI and 75% as Islands, Northwest Territories, Canada. Arctic 42:3–23. many as in the entire STI (Miller, 1993: BIR, 84 000 vs. GRACE, E.S. 1976. Interactions between men and wolves at an QEI, 20 500; and BIR, 84 000 vs. STI, 111 300)? arctic outpost on Ellesmere Island. Canadian Field-Naturalist 90:149– 156. (5) Is the potential ungulate prey base so low on the south- GRAY, D.R. 1993. The use of muskox kill sites as temporary central islands of the STI that it does not support any rendezvous sites by arctic wolves with pups in early winter. wolves on a year-round basis, as the questionnaire data Arctic 46:324–330. suggest (Miller, 1993: only 300 ungulates estimated MANNING, T.H. 1943. Notes on the of south and central within the entire area)? west Baffin Island. Journal of Mammalogy 24:47–59. MANNING, T.H., and MacPHERSON, A.H. 1958. The mammals (6) How would these results have been influenced if other of Banks Island. Arctic Institute of North America Technical researchers, who are known to have information on wolf- Paper No. 2. 74 p. sightings on Arctic Islands, had responded? MANNING, T.H., and MacPHERSON, A.H. 1961. A biological investigation of Prince of , N.W.T. Transactions of the Royal Canadian Institute 33:116– 239. ACKNOWLEDGEMENTS MILLER, F.L. 1978. Interactions between men, dogs and wolves on western Queen Elizabeth Islands, Northwest Territories, The study was supported by the Canadian Wildlife Service Canada. Musk-ox 22:70–72. (CWS), Environment Canada, and Polar Continental Shelf Project ———. 1990. Peary caribou status report. Environment Canada (PCSP), . We thank Bonni Hyrcyk, Report prepared for The Committee on the Status of Endangered Director, PCSP, for her continued support and for having the Wildlife in Canada. Edmonton, Alberta: Canadian Wildlife mailing list of arctic researchers supplied to us; S.J. Barry, CWS, for Service, Western and Northern Region. 64 p. statistical assistance; and J. Kennedy for drafting the maps. We ———. 1993. Status of wolves in the Canadian Arctic Archipelago. offer a special thanks to all of the 107 people who took the time and Technical Report Series No. 173. Edmonton, Alberta: Canadian made the effort to help us in our endeavour by responding to the Wildlife Service, Western and Northern Region. 63 p. wolf-sighting questionnaire. We are grateful to an anonymous MILLER, F.L., RUSSELL, R.H., and GUNN, A. 1977. Distributions, reviewer for providing a most helpful review of the original movements and numbers of Peary caribou and muskoxen on manuscript. western Queen Elizabeth Islands, Northwest Territories, 1972– 74. Canadian Wildlife Service Report Series No. 40. 55 p. MUNTHE, K., and HUTCHINSON, J.H. 1978. A wolf-human REFERENCES encounter on Ellesmere Island, Canada. Journal of Mammalogy 59:876– 878. ANDERSON, R.M. 1943. Summary of the large wolves of Canada, NORTHWEST TERRITORIES DATA BOOK 1990/91. 1990. with description of three new arctic races. Journal of Mammalogy Yellowknife: Outcrop Publishing Ltd. 238 p. 24:386– 393. PACKARD, J.M., MECH, L.D., and REAM, R.R. 1992. Weaning ———. 1946. Catalogue of Canadian recent mammals. National in an pack: Behavioral mechanisms. Canadian Journal Museum of Canada Bulletin 102 (Biological Series No. 31). of Zoology 70:1269–1275. Ottawa: Department of Mines and Resources Canada, Mines POCOCK, R.I. 1935. The races of Canis lupus. Proceedings of the and Geological Branch. 238 p. Zoological Society of Pt. 3:647–686. BABB, T.A., and BLISS, L.C. 1974. Susceptibility to environmental RIEWE, R., ed. 1992. atlas. Edmonton, Alberta: Canadian impact in the Queen Elizabeth Islands. Arctic 27:234–237. Circumpolar Institute and the Tungavik Federation of Nunavut. BLISS, L.C. 1990. High Arctic ecosystems: How they develop and 259 p. are maintained. In: Harington, C.R., ed. Canada’s missing TENER, J.S. 1963. Queen Elizabeth Islands game survey, 1961. dimension—science and history in the Canadian Arctic Islands, Canadian Wildlife Service Occasional Paper No. 6. 50 p. Volume 1. Ottawa: Canadian Museum of Nature. 350–384.